Evidence based practice (EBP) has become an aversive phrase for me over the past few years, even as I immersed myself in writing up a thesis for my PhD.
It’s a little surprising to reflect on this actually, considering one of the significant factors influencing my decision to undertake a PhD was that I love nothing more than immersing myself in the research. I started a PhD so I could have legitimate reason to sit on my but and read publications all day.
So how’d we go from loving research to being allergic to the phrase EBP?
Well the above myth busters article published in Feb 2021 in the journal Current Opinions in Pediatrics pretty much sums it up for me.
EBP is really just a term these days used to represent a phenomenon more accurately described as Publication Based Practice (PBP).
A one-legged stool that’s as flimsy as the recommendations from this publication.
The mythbuster publication looks at several hot topics in childhood development related to orthopedic concerns,
w-sitting, flexible flat foot, in/out-toeing and toe-walking
Hot topics because there is controversy around what, if and when to treat. Luckily this mythbuster publication answers all your complex questions and lays the controversy to rest.
IF you're practicing PBP.
I am not!
So just for fun I took a deep dive into this article to see how their recommendations aligned with clinical expertise – the most neglected of the 3 legs of true EBB. I only covered flat feet and w-sitting. The research around ITW similarly misrepresents this incredibly complex topic from a clinical perspective. I'll leave that for another day.
Let’s get into it...
The flexible flat foot
The myth buster article above cites 8 publications on the flexible flat foot to back up it’s ‘opinion’. This opinion which is frequently mistaken for fact by the PBP practitioner because it is authoritative (and lets face it), convenient to adhere to publications that dismiss the need for expert clinical reasoning in favor of simplified recommendations that are easy to apply and make it easier to manage those big wait-lists!
The myth buster article describes the flexible flat foot as an absence of the arch and a valgus heel on weight-bearing that corrects with non-weight bearing, toe-standing or big toe dorsiflexion. Recommendations from the 8 publications are essentially,
They'll grow out of it
Therefore you do not need to do anything, including prescribing custom or off-the-shelf orthotics. You could provide supportive shoes for cosmetic reasons (predominantly to appease parents) and you should refer to orthopaedics if feet are painful, stiff, ‘atypical’ or the child has a ‘functional deficit’. Pretty easy recommendations to follow right?
How are these recommendations supported by the 8 cited publications used to bust these myths?
Justification begins with a 1957 publication of 769 children in an essentially cross-sectional study of children from early walking age to age 11 (some children were used at two or three different time points)[i]. The study found that a valgus heel was strongly associated with ‘knock knees’ but that by age 7, only 2% of the population had ‘knock-knees’ indicating that 98% children would outgrow flat feet and knock knees. In other words, flat feet is a normal variant and 'they'll outgrow it'.
Well if you’ve heard the term lies, lies, lies and then statistics, this is a great example. Saying 98% of children outgrow something as justification for non-treatment is outrageous. If you’re not outraged, then let me put it in perspective for you. Cerebral Palsy is considered one of the most common motor disabilities in childhood. Estimates range between 1 – 4 children diagnosed with CP for every 1000 births (https://www.cdc.gov/ncbddd/cp/data.html). That’s 0.4% of the population. Imagine saying since 99.6% of children born do not have CP, we don’t need to provide treatment? Outrageous!
Anyway, lets jump 30 years forward and look at the next publication these mythbusting authors quoted from 1987[ii]. This was another cross sectional study of 882 feet in children and adults (aged 1-80 years). This publication looked at chalk footprints made by these feet on a paper and provided a range of normal from this data that did not require treatment, but noted that management outside these ranges was controversial. Of note, no musculoskeletal abnormalities were reported in their population, but they did not ask any of the owners of the 882 feet that fell in or outside the normal range about pain or dysfunction. Nor did they look at calcaneal valgus. So again – we have data that there is a range of flat to arched feet – some people may fall outside of this range and we should just ignore them as they are a small subset of the population - statistically inconsequential.
The next quoted publication was two years later and reported on a study that followed up a group of children for 3 years who were given either heel-cups, molded insoles, corrective shoes or no treatment[iii]. No matter the treatment, most marginally improved, even those children with no treatment, again justifying a no treatment approach. Calcaneal valgus was not used as a criteria for this study.
And of course most is not all.
We don’t know what the difference was in those that didn’t improve spontaneously and those that did.
But it’s really not a problem because as the authors of this 1989 publication wrote, since they've ‘provided the evidence' that you don’t actually need to treat flat feet, they have applied this in their clinic. Their standard approach when one or both parents come to their clinic with severe, symptomatic flat feet seeking treatment to prevent this future for their own children, is to do nothing. They confidently reassure parents that the child may indeed have a similar condition when he or she is fully grown, but that it can be treated with a supportive inserts at that time.
Wait – you mean there are adults who didn’t outgrow flat feet and now have severe, symptomatic flat feet?
We will just ignore that inconvenient fact and jump to the next myth busting study which takes place in 2007. This RCT compared custom orthoses to off-the-shelf inserts and no treatment[iv]. 160 children age 7 – 11 with flat feet (who had not outgrown their flat feet) were followed up for 12 months. These children showed a trend for improvement, regardless of treatment type or no treatment.
Children with developmental delay, motor coordination difficulties or any other neuromotor issues were excluded.
Now this is a good study. Criteria for an over pronated foot is well defined in the study and (in RCT land) a follow up of 12 months is excellent.
The authors defined flat feet as
excess pronation measured as calcaneal eversion of 5 degrees or more in resting calcaneal stance position (RCSP = the subject standing in a resting stance position in their preferred angle and base of gait i.e. their preferred standing position), and navicular drop of 10 mm or more.
Unsurprisingly, this 2007 study found that providing orthotics at 7 years of age did not ‘fix’ the foot or the foot pain.
This should be worrying you a little. Consider the previous advice to parents that all that was needed to treat adults with severe, symptomatic flat feet was supportive inserts. Ignoring the more nuanced texts and contradictions across studies and clinical practice in favor of a clean and simple clinical recommendation - do nothing.
And in case you missed it,
7-year-old children were presenting to the clinic with foot pain and flat feet. These are the 2% that had not 'outgrown it'.
Another prevalence study the myth busing authors quoted[v], described a pathological flat foot as one having a valgus position of >20 degrees or a valgus position that was not correctable (much higher cut-off than the 5% valgus of the above study. This prevalence study found 1% of preschool children fell into this range. That's huge! This shows that at least a whopping 1% of the population are going to be significantly disadvantaged by a wait and see approach.
(keep in mind that children with CP make up 0.4% so 1% is significant)
Is it possible that looking more specifically at calcaneal valgus, rather than broadly speaking about flat feet might start to give us more nuanced information?
The authors of the mythbuster publication choose to ignore this inconvenience as well as the information in a 2011 Cochrane review[vi] they reference. This Cochrane review noted that there is no standardized framework to evaluate the pediatric flat foot, no longitudinal studies looking at pain and dysfunction in adults, an uncertain prognosis and only 2 RCTs looking at pain.
It's much less complicated to ignore the nuances and in favor of the narrative that the paediatric flat foot is often unnecessarily treated and therefore we should not treat any.
I want to reflect just for a moment on the fact that the 2011 Cochrane review noted no standardized way to evaluate flat feet. And none of the 'evidence' quoted in the myth busting publication are post 2011.
This is a common conundrum in PBP. Pears are compared with apples and results are typically inconclusive, with more recent publications primarily consisting of systematic reviews of old research that level out any nuances to draw broad conclusions. The most authors can truthfully say with these systematic reviews is that more research is needed. And that was the most commonly stated conclusion we have seen in Cochrane reviews of rehabilitation science over the past many years. However there is a growing trend to push for clinical recommendations from publications. Authors make leaps in their clinical recommendations that should never be acceptable in good science. These leaps are quoted in summaries and abstracts and become the 'conclusion'.
Abstracts are often the only thing clinician's have access to.
And recommendations to 'do nothing' are easy to implement.
No wonder PBP is becoming the norm. It certainly take the pressure off having to take years to develop your clinical understanding of the underlying issues and possible solutions.
Let's take a look then at the last two publications use to support this myth busting publication. These were the most recent evidence used to justify the 'they outgrow it' conclusion. Both publications were published in 2020 by Choi et al. One was a systematic review[vii] which found no good evidence for the use of orthotics in asymptomatic flat feet.
We have to agree here. There is no good evidence.
However does that mean we should just ignore 1 – 2% of the population who are going to grow up to have severe, symptomatic flat feet?
Or does it mean that the evidence is not sufficient yet to inform practitioners about what works, when to intervene or how to differentiate between a flat foot that will be asymptomatic and one that will not?
In Choi et al's 2020 systematic review NO studies using hindfoot alignment radiographs were found. This lead the authors to suggest that
“Since a valgus deviation of the hindfoot alignment is an important patho-anatomical factor of the flat foot deformity, changes in the hindfoot alignment must be evaluated in studies that focus on the long-term effects of orthoses.”
The 2020 systematic review concluded (as do most systematic reviews in the field of neurodevelopment) that
“Due to the deficiency of related studies, heterogeneity, and low methodological quality of the studies within this review, definite conclusions could not be made”
Choi et al then published a prospective comparative study in 2020 of children aged 10+ who were prescribed an arch support, heel cup or no treatment[viii]. As with the 7 year olds in the 2007 study mentioned above, treatment or no treatment did not change the foot structure or the pain of these 10 year old significantly.
As with the 2007 study, in 2020 children were presenting with foot pain. And children had not outgrown their flat feet and providing orthotics in older children did not seem to be effective.
The 2020 systematic review and prospective study, along with the 2007 study certainly seem to answer one question at least,
It is not effective to remediate the painful flexible flat foot after the age 7 with an orthosis.
It would appear that waiting to treat the flexible flat foot is not a great idea and begs the question,
Is early intervention something worth looking at?
But that is not the question they were trying to answer, and so the myth busters gloss over it in favor of providing a clear clinical guideline. This is PBP.
PBP lacks critical analysis of the clinical problem, cherry picks outcomes that suit the narrative and relies heavily on old publications that are hashed and re-hashed in systematic reviews and reviews of reviews. This makes the evidence (aka publications) seem current and authoritative when in fact, the research is old, scant, inconclusive, with questionable standardization of definitions and measures and missing large populations of the children we are actually seeing in clinic.
Those children with motor control difficulties, development delays and more specific neuromotor issues were specifically excluded in most publications and the 2% of the population that are very likely to NOT outgrow their over pronation and might be on track to become those adults with severe, symptomatic flat feet were ignored.
And if you’re still with me, I’ll just quickly jump to the “evidence” for the ‘myth’ busting W-sitting recommendations. It won’t take long.
The myth busting authors note that the w-sitting myth is that w-sitting causes hip dysplasia, weakens the core and limits the ability to use both arms, leading to problems with coordination, gross motor skills and hand preferences amongst other things. As with flat feet, they go on to describe the prevalence of w-sitting in the general population to make the point that it is a variant of normal and therefore they will 'outgrow it'.
However they do temper this recommendation by noting moderation is a good idea so encourage children to use a variety of sitting (Clearly they've never worked with a persistent w-sitter) and refer to the orthopaedics if you have concerns about functional deficits, developmental delay or neuromuscular conditions.
There is so much wrong with their summary I could write a book. It’s a gross misrepresentation of current clinical thinking regarding w-sitting. If you want a true evidence based understanding of w-sitting that is grounded in excellent research and clinical analysis then go HERE
The myth busting authors use 7 references to make their point. Of these, 3 looked at prevalence, 2 looked at natural history of femoral anteversion and 1 was an orthopaedic textbook reference. The 7th publication compared hip x-rays to a parent’s report of their child’s current and past w-sitting behaviour[ix]. This publication found that in the 25 children, aged between 1 to 27 years (yes you read that correctly - 27 YEARS) who w-sat in the past or currently w-sat, compared to 79 'children' who did not or had never w-sat, there was no association between w-sitting and hip dysplasia. Apparently, this is sufficient to dispel any myth that w-sitting may impact development on any level.
How we make the leap from no association with hip dysplasia to dismissing ALL current clinical concerns about children who persistently w-sit neatly sums up my aversion to PBP proponents who quote this drivel as 'evidence based' on nothing more than the fact that it’s been published in a journal. It would be mildly amusing, if it wasn’t so concerning. The end point of repeating these half-truths and statistical facades without addressing the clinical and patient experience (the other two legs of the EBP stool) is that funders and insurers find it particularly helpful.
In summary then, I have to recognize that my aversion is not to EBP, but to PBP. Evidence based practice is important. It is important to be up to date with the evidence. To understand the natural history, the research findings and normal variants. To clinically appraise these in conjunction with the clinical and patient experience. Certainly treating children unnecessarily has many negative consequences, and promoting outdated beliefs or therapies may be harmful. But there is no excuse for PBP dressed up as EBP. PBP is lazy and contrary to the intent of true EBP that seeks clinically meaningful insight not a simplified checklist.
Lets reclaim EBP, and call out PBP when we see it
[i] Morley A. Knock-knee in children. Br Med J 1957; 2:976–979 [ii] Staheli LT, Chew DE, Corbett M. The longitudinal arch. A survey of eight hundred and eighty-two feet in normal children and adults. J Bone Jt Surg - Ser A 1987 [iii] Wenger DR, Mauldin D, Speck G, et al. Corrective shoes and inserts as treatment for flexible flatfoot in infants and children. J Bone Jt Surg Am 1989; 71:800 – 810. [iv] Whitford D, Esterman A. A randomized controlled trial of two types of in-shoe orthoses in children with flexible excess pronation of the feet. Foot Ankle Int 2007; 28:715 – 723 [v] Pfeiffer M, Kotz R, Ledl T, et al. Prevalence of flat foot in preschool-aged children. Pediatrics 2006; 118:634–639 [vi] Evans AM, Rome K. A Cochrane review of the evidence for nonsurgical interventions for flexible pediatric flat feet. Eur J Phys Rehabil Med 2011; 47:69 – 89 [vii] Choi JY, Hong WH, Suh JS, et al. The long-term structural effect of orthoses for & pediatric flexible flat foot: a systematic review. Foot Ankle Surg 2020; 26:181–188 [viii] Choi JY, Lee DJ, Kim SJ, Suh JS. Does the long-term use of medial arch & support insole induce the radiographic structural changes for pediatric flexible flat foot? A prospective comparative study. Foot Ankle Surg 2020;26:449 – 456. [ix] Rethlefsen SA, Mueske NM, Nazareth A, et al. Hip dysplasia is not more common in W-Sitters. Clin Pediatr (Phila). 2020